Mind-reading AI to translate silent speech into words

About this project 

This project aims to develop an AI system that can translate certain thoughts, such as silent speech, into text. By using non-invasive brainwave measurements, it will overcome the current impediment of easily translating silent speech into words, which would benefit an extensive range of Australian end-users.

The project aims to develop a system that can translate words that are not spoken aloud into speech for people to communicate and interact through their thoughts. It proposes an unprecedented model to process words in sentences to produce natural language. The system will adapt to individuals. Expected outcomes include new understanding of how the brain processes language, artificial intelligence (AI) models for interpreting data from brains and recognising speech elements, and a novel online feedback system to improve how humans and AI interact.

Applicants are required to have: 

• Strong publication record in the AI field, preferred in first-author CORE A*/A conferences.
• Proficiency in Python, machine learning frameworks (PyTorch/TensorFlow), and NLP tools (Hugging Face, SpaCy) 

Additionally, applicants with the following will be highly regarded:

• Strong background in neuroscience, artificial intelligence (AI), and natural language processing (NLP). 
• Skills in brain signal processing (EEG), deep learning (transformers, contrastive learning), and speech decoding. 
• Experience with multimodal AI, reinforcement learning for adaptive systems, and real-time feedback mechanisms is essential. 
• Should be innovative, capable of interdisciplinary collaboration, and interested in the ethical implications of mind-reading AI.

What you’ll do 

Promotion of the research outcomes will be achieved through various means. Firstly, open-source publication of the AI model and source code will allow the general public and industry to test and adopt the results, paving the path for technological and social adoption. Secondly, public-facing workshops, mainstream media publications and industry seminars will further assist potential translation, especially by collaboration with the fast-growing Australian assistive technologies industry sector. This is expected to result in job creation by enhancing Australia’s technological and commercial capacity, as well as socially contribute to safer, more effective and economically viable AI technologies ‘made in Australia’.

Where you’ll be based 

Industrial AI (IAI)

Dr Jimmy Cao, an ARC DECRA Fellow and UniSA Enterprise Fellow, will lead the project with his expertise in brain-computer interface and reinforcement learning. 

Professor Lin Liu will support the project through robust machine learning and time-series data mining.

Distinguished Professor Chin-Teng Lin will support this project through brain information processing and communication with machines.

Dr Thomas Do will support this project through human-computer interaction and robotic systems. Financial Support  

This project is funded for reasonable research expenses. Additionally, a living allowance scholarship of $35,200 per annum is available to eligible applicants. Australian Aboriginal and/or Torres Strait Islander applicants will be eligible to receive an increased stipend rate of $52,352 per annum (2025 rate). A fee-offset or waiver for the standard term of the program is also included. For full terms and benefits of the scholarship please refer to our scholarship information for domestic students or international students.

Eligibility and Selection

This project is open to applications from both Domestic and International applicants.
Applicants must meet the eligibility criteria for entrance into a PhD. 

Additionally, applicants must meet the project selection criteria: 

•    Strong publication record in the AI field, preferred in first-author CORE A*/A conferences.
•    Proficiency in Python, machine learning frameworks (PyTorch/TensorFlow), and NLP tools (Hugging Face, SpaCy) 

Applicants who can also demonstrate the following will be highly regarded:

•    Strong background in neuroscience, artificial intelligence (AI), and natural language processing (NLP). 
•    Skills in brain signal processing (EEG), deep learning (transformers, contrastive learning), and speech decoding. 
•    Experience with multimodal AI, reinforcement learning for adaptive systems, and real-time feedback mechanisms is essential. 
•    Should be innovative, capable of interdisciplinary collaboration, and interested in the ethical implications of mind-reading AI.

All applications that meet the eligibility and selection criteria will be considered for this project. A merit selection process will be used to determine the successful candidate.

The successful applicant is expected to study full-time and to be based at our Mawson Lakes Campus in the north of Adelaide. 

Essential Dates 

Applicants are expected to start in a timely fashion upon receipt of an offer. Extended deferral periods are not available. Applications close on Wednesday 11 June 2025.